Testing implications of the network theory of mental disorders in intensive short-term treatment

Importance. Network theory is increasingly used to inform targeted and personalized treatment by identifying symptoms and interactions that maintain psychopathology. Yet, its prediction that successful treatment should weaken symptom interactions remains insufficiently tested.Objective. To examine whether temporal interactions among depression and anxiety symptoms and global network connectivity decrease during short, intensive inpatient/day-clinic treatment, as predicted by network theory.Design. Preregistered secondary analysis of OPTIMA, a randomized controlled trial conducted September 2017 to October 2020; analyses were conducted January 2025 to May 2026.Setting. Single-center inpatient and day-clinic treatment in Germany with weekly symptom assessments.Participants. Adults aged 18 to 75 years with moderate to severe depression without psychotic symptoms. Of 1370 patients assessed, 292 were randomized, and 288 were included in this analysis.Interventions. Seven weeks of schema therapy, cognitive behavioral therapy, or individual supportive therapy, with up to 2 individual and 2 group sessions weekly plus clinically indicated pharmacotherapy and concomitant care.Main Outcomes and Measures: Changes in temporal interactions among 6 core depression and anxiety symptoms and in global network connectivity, estimated using a time-varying longitudinal network model. Exploratory stability landscapes characterized changes in network-implied symptom states.Results. Among 288 participants (mean [SD] age, 43.6 [13.6] years; 55.6% female), only 3 of 36 temporal symptom associations changed substantially, although symptom severity decreased. Mean absolute network strength did not substantially decrease from 0.55 (95% CI, 0.47-0.64) at treatment start to 0.46 (95% CI, 0.38-0.54) after 7 weeks. Still, the network model could explain symptom trajectories well (Bayesian pseudo-R² ranging from 0.43 to 0.54) and exploratory network-implied stability landscapes indicated a system shift toward lower-symptom states.Conclusions and Relevance. These findings provide limited support for the prediction that successful treatment reduces global connectivity in short, intensive treatment settings. In such settings, treatment-related change in symptom severity may be reflected in a reduction in general likelihood of symptom occurrence, but not necessarily in changes in symptom interactions, raising questions about long-term vulnerability. On the other hand, global symptom connectivity alone may not sufficiently characterize a healthy symptom system, and the role of specific symptom interactions warrants further investigation.Trial registration: ClinicalTrials.gov Identifier: NCT03287362.